Release 4.11 net/rds/tcp_listen.c

/*
 * Copyright (c) 2006 Oracle.  All rights reserved.
 *
 * This software is available to you under a choice of one of two
 * licenses.  You may choose to be licensed under the terms of the GNU
 * General Public License (GPL) Version 2, available from the file
 * COPYING in the main directory of this source tree, or the
 * OpenIB.org BSD license below:
 *
 *     Redistribution and use in source and binary forms, with or
 *     without modification, are permitted provided that the following
 *     conditions are met:
 *
 *      - Redistributions of source code must retain the above
 *        copyright notice, this list of conditions and the following
 *        disclaimer.
 *
 *      - Redistributions in binary form must reproduce the above
 *        copyright notice, this list of conditions and the following
 *        disclaimer in the documentation and/or other materials
 *        provided with the distribution.
 *
 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 * SOFTWARE.
 *
 */
#include <linux/kernel.h>
#include <linux/gfp.h>
#include <linux/in.h>
#include <net/tcp.h>

#include "rds.h"
#include "tcp.h"



int rds_tcp_keepalive(struct socket *sock)
{
	/* values below based on xs_udp_default_timeout */
	int keepidle = 5; /* send a probe 'keepidle' secs after last data */
	int keepcnt = 5; /* number of unack'ed probes before declaring dead */
	int keepalive = 1;
	int ret = 0;

	ret = kernel_setsockopt(sock, SOL_SOCKET, SO_KEEPALIVE,
				(char *)&keepalive, sizeof(keepalive));
	if (ret < 0)
		goto bail;

	ret = kernel_setsockopt(sock, IPPROTO_TCP, TCP_KEEPCNT,
				(char *)&keepcnt, sizeof(keepcnt));
	if (ret < 0)
		goto bail;

	ret = kernel_setsockopt(sock, IPPROTO_TCP, TCP_KEEPIDLE,
				(char *)&keepidle, sizeof(keepidle));
	if (ret < 0)
		goto bail;

	/* KEEPINTVL is the interval between successive probes. We follow
         * the model in xs_tcp_finish_connecting() and re-use keepidle.
         */
	ret = kernel_setsockopt(sock, IPPROTO_TCP, TCP_KEEPINTVL,
				(char *)&keepidle, sizeof(keepidle));
bail:
	return ret;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Sowmini Varadhan
158
100.00%
1
100.00%
Total
158
100.00%
1
100.00%

/* rds_tcp_accept_one_path(): if accepting on cp_index > 0, make sure the
 * client's ipaddr < server's ipaddr. Otherwise, close the accepted
 * socket and force a reconneect from smaller -> larger ip addr. The reason
 * we special case cp_index 0 is to allow the rds probe ping itself to itself
 * get through efficiently.
 * Since reconnects are only initiated from the node with the numerically
 * smaller ip address, we recycle conns in RDS_CONN_ERROR on the passive side
 * by moving them to CONNECTING in this function.
 */

static

struct rds_tcp_connection *rds_tcp_accept_one_path(struct rds_connection *conn)
{
	int i;
	bool peer_is_smaller = (conn->c_faddr < conn->c_laddr);
	int npaths = max_t(int, 1, conn->c_npaths);

	/* for mprds, all paths MUST be initiated by the peer
         * with the smaller address.
         */
	if (!peer_is_smaller) {
		/* Make sure we initiate at least one path if this
                 * has not already been done; rds_start_mprds() will
                 * take care of additional paths, if necessary.
                 */
		if (npaths == 1)
			rds_conn_path_connect_if_down(&conn->c_path[0]);
		return NULL;
	}

	for (i = 0; i < npaths; i++) {
		struct rds_conn_path *cp = &conn->c_path[i];

		if (rds_conn_path_transition(cp, RDS_CONN_DOWN,
					     RDS_CONN_CONNECTING) ||
		    rds_conn_path_transition(cp, RDS_CONN_ERROR,
					     RDS_CONN_CONNECTING)) {
			return cp->cp_transport_data;
		}
	}
	return NULL;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Sowmini Varadhan
129
99.23%
3
75.00%
Santosh Shilimkar
1
0.77%
1
25.00%
Total
130
100.00%
4
100.00%



int rds_tcp_accept_one(struct socket *sock)
{
	struct socket *new_sock = NULL;
	struct rds_connection *conn;
	int ret;
	struct inet_sock *inet;
	struct rds_tcp_connection *rs_tcp = NULL;
	int conn_state;
	struct rds_conn_path *cp;

	if (!sock) /* module unload or netns delete in progress */
		return -ENETUNREACH;

	ret = sock_create_kern(sock_net(sock->sk), sock->sk->sk_family,
			       sock->sk->sk_type, sock->sk->sk_protocol,
			       &new_sock);
	if (ret)
		goto out;

	new_sock->type = sock->type;
	new_sock->ops = sock->ops;
	ret = sock->ops->accept(sock, new_sock, O_NONBLOCK, true);
	if (ret < 0)
		goto out;

	ret = rds_tcp_keepalive(new_sock);
	if (ret < 0)
		goto out;

	rds_tcp_tune(new_sock);

	inet = inet_sk(new_sock->sk);

	rdsdebug("accepted tcp %pI4:%u -> %pI4:%u\n",
		 &inet->inet_saddr, ntohs(inet->inet_sport),
		 &inet->inet_daddr, ntohs(inet->inet_dport));

	conn = rds_conn_create(sock_net(sock->sk),
			       inet->inet_saddr, inet->inet_daddr,
			       &rds_tcp_transport, GFP_KERNEL);
	if (IS_ERR(conn)) {
		ret = PTR_ERR(conn);
		goto out;
	}
	/* An incoming SYN request came in, and TCP just accepted it.
         *
         * If the client reboots, this conn will need to be cleaned up.
         * rds_tcp_state_change() will do that cleanup
         */
	rs_tcp = rds_tcp_accept_one_path(conn);
	if (!rs_tcp)
		goto rst_nsk;
	mutex_lock(&rs_tcp->t_conn_path_lock);
	cp = rs_tcp->t_cpath;
	conn_state = rds_conn_path_state(cp);
	WARN_ON(conn_state == RDS_CONN_UP);
	if (conn_state != RDS_CONN_CONNECTING && conn_state != RDS_CONN_ERROR)
		goto rst_nsk;
	if (rs_tcp->t_sock) {
		/* Need to resolve a duelling SYN between peers.
                 * We have an outstanding SYN to this peer, which may
                 * potentially have transitioned to the RDS_CONN_UP state,
                 * so we must quiesce any send threads before resetting
                 * c_transport_data.
                 */
		if (ntohl(inet->inet_saddr) < ntohl(inet->inet_daddr) ||
		    !cp->cp_outgoing) {
			goto rst_nsk;
		} else {
			rds_tcp_reset_callbacks(new_sock, cp);
			cp->cp_outgoing = 0;
			/* rds_connect_path_complete() marks RDS_CONN_UP */
			rds_connect_path_complete(cp, RDS_CONN_RESETTING);
		}
	} else {
		rds_tcp_set_callbacks(new_sock, cp);
		rds_connect_path_complete(cp, RDS_CONN_CONNECTING);
	}
	new_sock = NULL;
	ret = 0;
	goto out;
rst_nsk:
	/* reset the newly returned accept sock and bail */
	kernel_sock_shutdown(new_sock, SHUT_RDWR);
	ret = 0;
out:
	if (rs_tcp)
		mutex_unlock(&rs_tcp->t_conn_path_lock);
	if (new_sock)
		sock_release(new_sock);
	return ret;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Andy Grover
212
50.00%
1
5.00%
Sowmini Varadhan
201
47.41%
16
80.00%
Eric Dumazet
6
1.42%
1
5.00%
Joe Perches
3
0.71%
1
5.00%
David Howells
2
0.47%
1
5.00%
Total
424
100.00%
20
100.00%



void rds_tcp_listen_data_ready(struct sock *sk)
{
	void (*ready)(struct sock *sk);

	rdsdebug("listen data ready sk %p\n", sk);

	read_lock_bh(&sk->sk_callback_lock);
	ready = sk->sk_user_data;
	if (!ready) { /* check for teardown race */
		ready = sk->sk_data_ready;
		goto out;
	}

	/*
         * ->sk_data_ready is also called for a newly established child socket
         * before it has been accepted and the accepter has set up their
         * data_ready.. we only want to queue listen work for our listening
         * socket
         *
         * (*ready)() may be null if we are racing with netns delete, and
         * the listen socket is being torn down.
         */
	if (sk->sk_state == TCP_LISTEN)
		rds_tcp_accept_work(sk);
	else
		ready = rds_tcp_listen_sock_def_readable(sock_net(sk));

out:
	read_unlock_bh(&sk->sk_callback_lock);
	if (ready)
		ready(sk);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Andy Grover
84
80.77%
2
33.33%
Sowmini Varadhan
18
17.31%
3
50.00%
Eric Dumazet
2
1.92%
1
16.67%
Total
104
100.00%
6
100.00%



struct socket *rds_tcp_listen_init(struct net *net)
{
	struct sockaddr_in sin;
	struct socket *sock = NULL;
	int ret;

	ret = sock_create_kern(net, PF_INET, SOCK_STREAM, IPPROTO_TCP, &sock);
	if (ret < 0)
		goto out;

	sock->sk->sk_reuse = SK_CAN_REUSE;
	rds_tcp_nonagle(sock);

	write_lock_bh(&sock->sk->sk_callback_lock);
	sock->sk->sk_user_data = sock->sk->sk_data_ready;
	sock->sk->sk_data_ready = rds_tcp_listen_data_ready;
	write_unlock_bh(&sock->sk->sk_callback_lock);

	sin.sin_family = PF_INET;
	sin.sin_addr.s_addr = (__force u32)htonl(INADDR_ANY);
	sin.sin_port = (__force u16)htons(RDS_TCP_PORT);

	ret = sock->ops->bind(sock, (struct sockaddr *)&sin, sizeof(sin));
	if (ret < 0)
		goto out;

	ret = sock->ops->listen(sock, 64);
	if (ret < 0)
		goto out;

	return sock;
out:
	if (sock)
		sock_release(sock);
	return NULL;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Andy Grover
195
92.86%
1
20.00%
Sowmini Varadhan
13
6.19%
2
40.00%
Pavel Emelyanov
1
0.48%
1
20.00%
Himangi Saraogi
1
0.48%
1
20.00%
Total
210
100.00%
5
100.00%



void rds_tcp_listen_stop(struct socket *sock, struct work_struct *acceptor)
{
	struct sock *sk;

	if (!sock)
		return;

	sk = sock->sk;

	/* serialize with and prevent further callbacks */
	lock_sock(sk);
	write_lock_bh(&sk->sk_callback_lock);
	if (sk->sk_user_data) {
		sk->sk_data_ready = sk->sk_user_data;
		sk->sk_user_data = NULL;
	}
	write_unlock_bh(&sk->sk_callback_lock);
	release_sock(sk);

	/* wait for accepts to stop and close the socket */
	flush_workqueue(rds_wq);
	flush_work(acceptor);
	sock_release(sock);
}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Andy Grover
85
87.63%
2
50.00%
Sowmini Varadhan
12
12.37%
2
50.00%
Total
97
100.00%
4
100.00%

Overall Contributors
Person
Tokens
Prop
Commits
CommitProp
Andy Grover
592
51.79%
2
6.67%
Sowmini Varadhan
532
46.54%
20
66.67%
Eric Dumazet
8
0.70%
2
6.67%
Joe Perches
3
0.26%
1
3.33%
Tejun Heo
3
0.26%
1
3.33%
David Howells
2
0.17%
1
3.33%
Himangi Saraogi
1
0.09%
1
3.33%
Pavel Emelyanov
1
0.09%
1
3.33%
Santosh Shilimkar
1
0.09%
1
3.33%
Total
1143
100.00%
30
100.00%
Directory: net/rds

Information contained on this website is for historical information purposes only and does not indicate or represent copyright ownership.